Commits

Breadcrumbs

History for

onroadmap-update

User selector

All users

All time

Commit History

Commits on Jan 24, 2017

• Remove concat_v2 from r1.0 branch

  

  annarev

  authored and

  gunan

  committed
  ba45e40

  Copy full SHA for ba45e40

Commits on Jan 23, 2017

• Fix: Fix timing flakiness by mocking time object in (#6944)

  Show description for 08d5fdb
  gunan

  committed
  08d5fdb

  Copy full SHA for 08d5fdb

Commits on Jan 20, 2017

• Fix: Bump the grace period in queue_runner_test.

  gunan

  committed
  3921ed2

  Copy full SHA for 3921ed2

• Default reuse should be None, never False. Correct it in layers and add tests.

  Show description for d8918bb
  

  Lukasz Kaiser

  authored and

  gunan

  committed
  d8918bb

  Copy full SHA for d8918bb

• Make "Neg" op hidden from Python.

  Show description for 510560e
  

  aselle

  authored and

  gunan

  committed
  510560e

  Copy full SHA for 510560e

• Change weight broadcasting for losses.sparse_softmax_cross_entropy to be the same as sparse_softmax_cross_entropy_with_logits. This means allowing greater than rank 2 inputs.

  Show description for 8a01c64
  

  tensorflower-gardener

  authored and

  gunan

  committed
  8a01c64

  Copy full SHA for 8a01c64

• Rename `weights` in `Dense` layer to `kernel`, and add base layer aliases for `(non_)trainable_weights`.

  Show description for aa4c188
  

  fchollet

  authored and

  gunan

  committed
  aa4c188

  Copy full SHA for aa4c188

• Make dtype consistent with docs for TP/TN/FP/FN ops.

  Show description for 5572e35
  

  tensorflower-gardener

  authored and

  gunan

  committed
  5572e35

  Copy full SHA for 5572e35

• Restrict weights rank to be the same as the broadcast target, to avoid ambiguity on broadcast rules.

  Show description for b731e37
  

  tensorflower-gardener

  authored and

  gunan

  committed
  b731e37

  Copy full SHA for b731e37

Commits on Jan 17, 2017

• Merge commit for internal changes

  

  Patrick Nguyen

  committed
  19ff874

  Copy full SHA for 19ff874

• Use integer division to compute number of updates.

  Show description for 0b56eab
  

  Patrick Nguyen

  authored and

  tensorflower-gardener

  committed
  0b56eab

  Copy full SHA for 0b56eab

• Remove tf.cloud from public API until it's ready.

  Show description for cddb86f
  

  Patrick Nguyen

  authored and

  tensorflower-gardener

  committed
  cddb86f

  Copy full SHA for cddb86f

Commits on Jan 16, 2017

• tfdbg: Add GrpcDebugWrapperSession

  Show description for 1580d3b
  

  caisq

  authored and

  tensorflower-gardener

  committed
  1580d3b

  Copy full SHA for 1580d3b

Commits on Jan 15, 2017

• Add tf.tables_initializer as a replacement for tf.initialize_all_tables, and

  Show description for 9654cf9
  tensorflower-gardener

  committed
  9654cf9

  Copy full SHA for 9654cf9

• Default naming of scopes should continue, not reset, after jumps to reused scopes.

  Show description for e720678
  

  Lukasz Kaiser

  authored and

  tensorflower-gardener

  committed
  e720678

  Copy full SHA for e720678

Commits on Jan 14, 2017

• Have `write_graph` return the output path of file (#6851)

  Show description for 48c433f
  

  samjabrahams

  authored and

  drpngx

  committed
  48c433f

  Copy full SHA for 48c433f

• Automated rollback of change 142694447

  Show description for 7ab67d8
  tensorflower-gardener

  committed
  7ab67d8

  Copy full SHA for 7ab67d8

• Update einsum to check whether uncompacted_shape has multiple None values (#6830)

  Show description for ca09a89
  

  jihunchoi

  authored and

  drpngx

  committed
  ca09a89

  Copy full SHA for ca09a89

Commits on Jan 13, 2017

• Fix documentation sample code (#6836)

  Show description for 6b4dd84
  

  normanheckscher

  authored and

  drpngx

  committed
  6b4dd84

  Copy full SHA for 6b4dd84

• Makes the ops used by ParallelConcat internal.

  Show description for b24a0b2
  tensorflower-gardener

  committed
  b24a0b2

  Copy full SHA for b24a0b2

• Do parallel_stack as a graph rewrite instead of python code.

  Show description for 705cc93
  tensorflower-gardener

  committed
  705cc93

  Copy full SHA for 705cc93

• Fix issues with shape inference during graph import.

  Show description for 8f89336
  

  petewarden

  authored and

  tensorflower-gardener

  committed
  8f89336

  Copy full SHA for 8f89336

• tfdbg: add Session.sess_str check to Session wrapper base class

  Show description for 7c96ead
  

  caisq

  authored and

  tensorflower-gardener

  committed
  7c96ead

  Copy full SHA for 7c96ead

• Make weight broadcasting more strict (in particular, catch the case where a 1-dim in values shape could be broadcast across a non-1 dimension in weights).

  Show description for 16cc887
  tensorflower-gardener

  committed
  16cc887

  Copy full SHA for 16cc887

• Fix reference to now-deleted function field.

  Show description for b6671e4
  tensorflower-gardener

  committed
  b6671e4

  Copy full SHA for b6671e4

• Make tfdbg tests pass in Windows Bazel build (#6821)

  

  caisq

  authored and

  drpngx

  committed
  018583b

  Copy full SHA for 018583b

• Merge changes from github.

  Show description for 769496b
  

  Patrick Nguyen

  authored and

  tensorflower-gardener

  committed
  769496b

  Copy full SHA for 769496b

• More new ops for speeding up input pipline: stage and unstage.

  Show description for 85acf52
  tensorflower-gardener

  committed
  85acf52

  Copy full SHA for 85acf52

• Create a benchmark for resize_bicubic_op.

  Show description for 221a8c6
  

  saeta

  authored and

  tensorflower-gardener

  committed
  221a8c6

  Copy full SHA for 221a8c6

• Remove unnecessary control dependencies.

  Show description for dac3cf8
  tensorflower-gardener

  committed
  dac3cf8

  Copy full SHA for dac3cf8

• Add nest.map_structure.

  Show description for 1013cee
  

  ebrevdo

  authored and

  tensorflower-gardener

  committed
  1013cee

  Copy full SHA for 1013cee

• Remove tf.mul, tf.neg and tf.sub from the public API

  Show description for 78b11ce
  

  aselle

  authored and

  tensorflower-gardener

  committed
  78b11ce

  Copy full SHA for 78b11ce

Commits on Jan 12, 2017

• Update docstring for SVD to emphasize that output argument order differs from numpy.linalg.svd.

  Show description for ef31cf3
  tensorflower-gardener

  committed
  ef31cf3

  Copy full SHA for ef31cf3

• Losses should support scalar logits/labels.

  Show description for 5ccc994
  tensorflower-gardener

  committed
  5ccc994

  Copy full SHA for 5ccc994

• Transpose confusion matrix output to make it consistent with sklearn.

  Show description for 4d9a240
  tensorflower-gardener

  committed
  4d9a240

  Copy full SHA for 4d9a240